© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 41244
    Prev
    Next
    INDEX
    Explanations

    proper nouns and names

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    gated
    Context Size
    1,024
    Dataset
    Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ellido
    -0.15
    indow
    -0.15
    arel
    -0.14
    abay
    -0.14
    usz
    -0.14
    ÑĢеж
    -0.14
    bert
    -0.14
    etailed
    -0.14
    alez
    -0.14
    acic
    -0.14
    POSITIVE LOGITS
     Le
    0.19
    frog
    0.17
    hana
    0.16
    nex
    0.15
    gue
    0.15
    @nate
    0.15
    (le
    0.15
    Normalization
    0.15
    она
    0.15
    rette
    0.14
    Activations Density 0.016%

    No Known Activations